Upload
doantruc
View
218
Download
0
Embed Size (px)
Citation preview
1
Davide ArcellaDavide Arcella
Data Collection and Exposure UnitData Collection and Exposure Unit
EFSAEFSA’’ssConcise European food Concise European food consumption databaseconsumption database
22
The EFSA raison d’être
Risk assessment authority created in 2002 as part of a comprehensive program to improve EU food safety, ensure a high level of consumer protection and restore and maintain confidence in the EU food supply.
33
EFSA task
In close collaboration with national authorities and in open consultation with its stakeholders, EFSA as the food safety risk assessor provides independent scientific advice and communication on existing and emerging risks.
44
Regulation (EC) N°178/2002
• EFSA “shall search for, collect, collate, analyse and summarise relevant scientific and technical data in the fields within its mission. This shall involve in particular the collection of data relating to food consumption and the exposure of individuals to risks related to the consumption of food”;
• EFSA “shall work in close cooperation with all organisations operating in the field of data collection, including those from applicant countries, third countries or international bodies”.
55
•• CollectionCollection, , collationcollation, and , and analysisanalysis of relevant scientific of relevant scientific
and technical data on and technical data on
–– food consumptionfood consumption
–– food compositionfood composition
–– food (and feed) contaminants food (and feed) contaminants
•• Support to EFSASupport to EFSA’’s Scientific panels and Scientific s Scientific panels and Scientific
Committee for their Committee for their exposure assessmentsexposure assessments and and
contribute to contribute to newnew exposure assessment methodologiesexposure assessment methodologies
DATEXDATEX’’s Missions Mission
66
Data storage frameworkData storage framework
Ad hoc occurrence
data collection
Concise database for consumption
data
Occurrence database
Consumption database
Exposure assessments
Additional food consumption data
“Official” and “commercial food composition dataFood
composition database
Maximum Limits
Maximum Limits (ML) database
Substance Dictionary
Food Dictionary
77
Food consumption data Food consumption data are an essential element for the are an essential element for the risk assessment work conducted by the EFSA.risk assessment work conducted by the EFSA.
Hence, the availability of Hence, the availability of reliablereliable and and detaileddetailed data in data in this domain is essential to enable it to carry out its this domain is essential to enable it to carry out its
mandates.mandates.
It is fundamental to take into special It is fundamental to take into special consideration consideration nonnon--average individualsaverage individuals, and in , and in
particular particular high consumershigh consumers (those who consumes (those who consumes relatively large quantities of foods). relatively large quantities of foods).
88
Pregnant women
Small children
Vegetarians, diabetics, …
Infants
Elderly
Consumption levels vary with age and physiological status
99
Consumption levels vary with geographical area
0
1
2
3
4
5
6
7
8
9
10
Bulg
aria
Italy
France
Latvi
aG
reec
eSw
eden
Portugal
Slove
niaR
oman
iaH
ungary
Estonia
Lithuan
iaFin
land
Spain
Net
herla
ndsPola
ndSlo
vaki
aB
elgiu
mD
enm
ark
Unite
d Kin
gdom
Luxem
bourgA
ustria
Ger
man
yIre
land
Cze
ch R
epublic
Average consumption of Average consumption of beer (L/capita per year)beer (L/capita per year) in the in the adult population in 25 EU Member States (WHO, 2007)adult population in 25 EU Member States (WHO, 2007)
1010
Dietary surveys among individualsDietary surveys among individuals
Main factors affecting reliability and completenessof the data:
Survey methodology (individual dietary record, 24-hour recall, …)
Number of days
Food codification level
Composite foods / recipes
Food consumption data from dietary surveys are available in a majority of European countries, but data obtained at national level can often not be compared directly because of various survey methodologies and various food categorisation systems.
11
Individual dietary record
219 – 644,2851997 – 98The Netherlands7
8
6
5
4
3
2
1
716 – 641,5441994 – 96Italy*
718 – 651,3791997 – 99Ireland
715 – 921,4741998 – 99France*
318 – 961,1792003 – 04Hungary
74 – 754,4392000 – 05Denmark
19 – 64
17 – 79
Age range
71,7242000 – 01United Kingdom
71,2101997 – 98Sweden
Number of days
Number of subjects
YearCountry
* New dietary survey just completed
12
24-hour recall
119 – 652,0151997Estonia5
225 – 652,0072002Finland6
119 – 652,0651993 – 97Austria1
117 – 602,2092006Slovakia10
11 – 964,1342000Poland9
119 – 652,0941997Lithuania8
317 – 601,0602002Spain11
7
4
3
2
116 – 648532004Bulgaria
216 – 641,7512003 – 04Czech Republic
216 – 641,7232004Belgium
16 – 64
Age range
11,3662002Iceland
Number of days
Number of subjects
YearCountry
13
Others
2817 – 794,0301998Germany*
(dietary-history)1
-15 – 792,3221997
Norway
(food frequency questionnaire)
2
Age
range
Number
of days
Number of
subjectsYearCountry
Latvia
Greece
2
1
Switzerland
Slovenia
6
5
Malta
Luxemburg
4
3 7
Portugal
Turkey
Data currently not available
* New dietary survey just completed
1414
“A common database on food consumption would improve the consistency and reliability of exposure assessments carried out by the various EFSA Panels and other experts in Europe”.
EFSA Scientific Colloquium 3
1515
EFSA Scientific Colloquium 3
Conclusions and recommendations
In the short term:“… the compilation of existing food consumption data can be performed in order to make data as comparable as possible across Europe”.
In the long term: “… harmonised food consumption data collection should be promoted in order to obtain a fully harmonised European database.
1616
Expert group on food consumption data
• To provide a platform for exchange of views with the best experts in Europe on ways to harmonise methodologies for the collection and collation of food consumption data
• To coordinate and facilitate the merger of national food consumption information into a pan-European food consumption database
Austria, Belgium, Bulgaria, Croatia, Cyprus, Czech Republic, Denmark, Estonia, Finland, France, Germany, Greece, Hungary, Iceland, Ireland, Italy, Latvia, Lithuania, Luxemburg, Malta, Norway, Poland, Portugal, Romania, Slovakia, Slovenia, Spain, Sweden, Switzerland, The Netherlands, Turkey and United Kingdom.
1717
Network meetingsNetwork meetings
11stst Expert Group network meeting held in January 2007 Expert Group network meeting held in January 2007
–– outlined activities in respective Member Stateoutlined activities in respective Member State
–– start the compilation of the Concise European Food start the compilation of the Concise European Food Consumption DatabaseConsumption Database
22ndnd Expert Group network meeting held in October 2007Expert Group network meeting held in October 2007
–– agreed to publish Member State aggregated agreed to publish Member State aggregated summaries on EFSA websitesummaries on EFSA website
–– Food Consumption and Exposure Working Group Food Consumption and Exposure Working Group
1818
Concise European food consumption database
The Concise European Food Consumption Database is called “concise” since it is intended to provide a limited number of data to be used as a screening tool for preliminary exposure assessments by the EFSA Scientific Panels and Member States.
It has been developed in order to allow risk managers to identify substances that might be of concern and prioritise the use of resources for safety assessments.
At the present time, this database is planned to contain food consumption data only for the adult population (16 to 64 years old).
1919
Each member of the Expert Group is in charge of coordinating the
– collection, – formatting and– transfer
of local National food consumption data to EFSA.
For this purpose, Expert Group members have been asked to:
– identify the most recent and relevant dietary survey available in his country and
– re-codify the food consumption database of the identified survey according to ad hoc broad food categories and subcategories.
Compilation of the Concise database
2020
To overcome the categorisation problem, EFSA has developed an ad hoc system with 15 main food categories (29 sub-categories).
The food categorisation system has been largely built by aggregating the Euro Food Grouping (EFG) categoriesdeveloped within the EFCOSUM project.
The number of food categories was limited in order to:
– increase comparability between countries and
– to allow the use of a conservative technique of exposure assessment.
– distinguish solid from liquid foods.
Food classification system
21
Food categories (1)
Fruit and vegetable juices7A
Soft drinks with percentage of fruits lower than nectar, excl fruit juice
7B
Vegetable soups4A
Cereal-based mixed dishes1A
Sub-categoriesMain food categories
Coffee, tea, cocoa (expressed as liquid)8
Bottled water7C
Fruit and vegetable juices, soft drinks and bottled water
7
Fruits6
Starchy roots or potatoes5
Vegetables, nuts, pulses except vegetable soups4B
Vegetables, nuts, pulses including carrots, tomato and leafy vegetables
4
Fats (vegetable and animal)3
Sugar & sugar products including chocolate2
Cereals & cereal products excl. Cereal-based mixed dishes
1BCereals & cereal products1
22
Food categories (2)
Eggs12
Milk and dairy based drinks13A
Milk and dairy based products13 Dairy based products13B
Cheese13C
Beer and substitutes9A
Alcoholic beverages9 Wine and substitutes9B
Other alcoholic beverages and substitutes9C
Meat based preparations10C
Edible offal and offal products10B
Meat and meat products and substitutes10A
Meat and meat products, offal10
Seafood and seafood products11A
Fish and fish products11B
Sub-categoriesMain food categories
Tap water15
Food for special dietary uses14B
Miscellaneous14AMiscellaneous / Food for special dietary uses
14
Fish based preparations11C
Fish and seafood11
2323
EFSA’s Concise European food consumption database
1) Belgium2) Bulgaria 3) Czech Republic4) Denmark5) Finland6) France7) Germany8) Great Britain9) Hungary10) Iceland11) Ireland12) Italy13) Norway14) Poland15) Slovakia16) Sweden17) The Netherlands
24
Format of the databases received
29%5Excel different from EFSA Template
17Total
10
2
Number of datasetsFormat
59%EFSA Template – Excel
12%EFSA Template – XML
2525
Summary statistics from the Concise database
Consumption data are submitted as average daily consumption per person.
Individual food consumption data are stored by EFSA but only summary statistics from the Concise database are made available to the public on the EFSA web site.
The use of food consumption data at individual level is restricted to EFSA staff and requires preliminary notification to the countries providing the data.
Any further use of the data at individual level requires a formal authorisation by each country providing the data.
2626
Data controls
– Unfilled food categories (all zeros and missing)
– Inconsistencies, e.g. Category 1 different from category 1B
but category 1A not provided.
– Missing values, e.g. missing values for the body weight (till
to 5.4% of the total population)
– Impossible values, e.g. body weight equal to “999” or “-4”
where considered as missing
– Possible outliers, e.g. energy intake > 1,000,000 KJoule
– Reference not reported
– Weighting factors
2727
Summary statistics from the Concise database
For each country, food consumption data are elaborated according to both main categories and sub-categories and for the total population and consumers only.
The summary statistics include:
– numbers of consumers,
– mean consumption,
– standard deviation,
– low and high percentiles of consumption.
3030
Use of the Concise Database
The Concise Database is intended to produce conservative estimates of exposure. If the database is used for screening assessments, an analysis of uncertainty is not required, provided that appropriate conservative assumptions take account of uncertainty.
Risk assessors are responsible for ensuring that the use of the database is conservative for the specific case. If data from the Concise Database are used for non-conservative assessments, the degree of uncertainty of the adopted model should be evaluated and discussed.
3131
Between countries differences
Between countries differences (especially in higher percentiles) can simply be induced by the survey methodology.
In particular, the duration of the survey is expected to affect the distribution of consumption, particularly at the upper tails.
A short survey tends to under-estimate the proportion of individuals who consume particular food types, but at the same time, over-estimates the levels for high consumers.
3232
High percentiles
The reliability of high percentiles is related to the number of subjects used to calculate them.
Percentiles calculated on a limited number of subjects bear a large uncertainty and only provide a rough indication of high levels of consumption.
The minimum sample sizes can be estimated
– between 130 and 160 for the 95th percentile,
– between 263 and 320 for the 97.5th percentile and
– between 662 and 800 for the 99th percentile.
3333
In addition to the food consumption data, a detailed description of the survey characteristics (e.g. method, duration, year, etc.) is provided in order to allow a correct interpretation of the data.
Information on the surveys
3434
Usage of the data from the Concise Database
Preliminary data:
Opinion of the EFSA Scientific Panel on Contaminants in the Food Chain on perfluorooctane sulfonate (PFOS), perfluorooctanoic acid (PFOA) and their salts.
Opinion of the EFSA Scientific Panel on Contaminants in the food chain related to ochratoxin A in food.
Under consideration:
EFSA scientific opinion on the risks to human health related to the presence of cadmium in foodstuffs.
Refinement of the 2001 COM report on food additivedietary intake at EU level..
3535
Short and long term objectives
In the short termIn the short term::
compilation of existing food consumption data compilation of existing food consumption data
•• at the lowest possible level of detailat the lowest possible level of detail
•• for population groups other than adults, e.g. small for population groups other than adults, e.g. small children, elderly, children, elderly, ……
In the long termIn the long term: :
Promote the collection of harmonised food consumption Promote the collection of harmonised food consumption data collectiondata collection
•• by Member Statesby Member States
•• within a panwithin a pan--European dietary surveyEuropean dietary survey
3636
Individual food consumption data for children
Deadline for submitting proposals is: 6th of June 2008.
3838
Food Consumption and Exposure Working Group (FCE WG)
• determine the feasibility of creating a Pan-European food consumption survey
• outline requirements on food consumption studies with respect to exposure assessment
• recommend methods for food consumption surveys with the view of better harmonisation across Europe,
• explore access to, translation and transfer of data from existing food consumption databases,
• investigate possible collaboration with other current European initiatives in the field (e.g. EFCOVAL project, EUROSTAT, IARC, etc.)
3939
EFCOSUM - European Food Consumption Survey Method projectEPIC - European Prospective Investigation into Cancer and NutritionEFCOVAL - European Food Consumption Validation projectFACET - Flavours, Additives and Contact exposure project
European research projects